README: v2 sol-refit release notes
Browse files
README.md
CHANGED
|
@@ -14,17 +14,27 @@ library_name: coreml
|
|
| 14 |
On-device replacement for the remote LLM that extracts structured numbers from
|
| 15 |
free-text exercise entries in the A-LIST app (locomo mode). Single fp16 Core ML
|
| 16 |
graph: MiniLM-L6 encoder (frozen, mean-pool + L2-norm inside the graph) + MLP
|
| 17 |
-
heads
|
| 18 |
|
| 19 |
- **Inputs**: `input_ids` int32 (1,128) WordPiece/bert-base-uncased; `attention_mask`
|
| 20 |
int32 (1,128); `parsed` float32 (1,10) regex features (see the app's
|
| 21 |
`ExerciseFeatureParser` for the exact order and transforms).
|
| 22 |
- **Output**: float32 (1,8) — `[log1p(duration_min), log1p(calories), heart_rate_bpm,
|
| 23 |
intensity_1_10, steps_gate_logit, log1p(steps), distance_gate_logit, log1p(distance_km)]`.
|
| 24 |
-
|
| 25 |
-
|
| 26 |
-
|
| 27 |
-
- `
|
| 28 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
|
| 30 |
Internal model for the A-LIST iOS app. Not intended for third-party use.
|
|
|
|
| 14 |
On-device replacement for the remote LLM that extracts structured numbers from
|
| 15 |
free-text exercise entries in the A-LIST app (locomo mode). Single fp16 Core ML
|
| 16 |
graph: MiniLM-L6 encoder (frozen, mean-pool + L2-norm inside the graph) + MLP
|
| 17 |
+
heads.
|
| 18 |
|
| 19 |
- **Inputs**: `input_ids` int32 (1,128) WordPiece/bert-base-uncased; `attention_mask`
|
| 20 |
int32 (1,128); `parsed` float32 (1,10) regex features (see the app's
|
| 21 |
`ExerciseFeatureParser` for the exact order and transforms).
|
| 22 |
- **Output**: float32 (1,8) — `[log1p(duration_min), log1p(calories), heart_rate_bpm,
|
| 23 |
intensity_1_10, steps_gate_logit, log1p(steps), distance_gate_logit, log1p(distance_km)]`.
|
| 24 |
+
|
| 25 |
+
## Versions
|
| 26 |
+
|
| 27 |
+
- **`v2/` (current, 2026-08-23)**: identical graph retrained on the sol-relabeled
|
| 28 |
+
dataset (exercise-text-unified:v0 — one uniform frontier teacher, calorie CV 1.6%
|
| 29 |
+
vs the old labels' ~11%). On the frozen held-out-user sol eval (1,127 entries) it
|
| 30 |
+
beats the v1 artifact on every field: duration 68.5% vs 62.7%, calories 58.0% vs
|
| 31 |
+
51.5%, heart rate ±10bpm 76.0% vs 67.6%, intensity ±1 81.7% vs 76.0%, steps gate
|
| 32 |
+
94.9% vs 88.6%, distance gate 94.6% vs 90.8%. Same I/O contract and footprint.
|
| 33 |
+
- **Root files (v1, 2026-08-16)**: distilled from ~14.5k historical production
|
| 34 |
+
extractions (mixed teacher generations). Kept resolvable for rollback; the app's
|
| 35 |
+
`ExerciseExtractorDescriptor.v1()` still pins these bytes.
|
| 36 |
+
- `golden_cases.json` holds 10 input→output cases generated from the v1 artifact —
|
| 37 |
+
the app uses them to pin the *decode rules* (weights-independent), so they remain
|
| 38 |
+
valid across versions.
|
| 39 |
|
| 40 |
Internal model for the A-LIST iOS app. Not intended for third-party use.
|